Little is known about what combinations of alcohol, drug abuse and mental health services are most appropriate for patients with co-occurring psychiatric and addictive disorders. Data from a pilot study at Beth Israel Medical Center demonstrated that schizophrenic substance abusers had a significant reduction in yearly average hospital days after treatment in a weekly outpatient group integrating psychiatric and substance abuse services. The specific aims of this study are: 1) To examine prospectively the effects of integrating psychiatric and addiction services in one location versus multi-site, multi-service deployment upon readmission of patients with schizophrenia and co- occurring addictive disorders to the hospital; 2) To demonstrate that integrated services for these patients are sufficient in reducing inpatient hospital days, although such services are not comprehensive; 3) To demonstrate that patients with schizophrenia and addictive disorders will maintain a connection to ongoing treatment if these service are temporally, physically, and philosophically integrated, and that such treatment will have positive effects both upon sobriety and psychiatric symptoms; 4) To provide longitudinal clinical and service-use data on these patients through the use of serial evaluations by trained raters and patient tracking in the community; 5) To provide a useful clinical model for treatment of severely mentally ill patients with alcohol and/or drug abuse disorders. This is a prospective, single-blind comparison study of two types of treatment for patients with diagnoses of DSMIIIR Schizophrenia and alcohol and/or other substance dependence. From index hospitalization, patients will be randomized into two outpatient groups; the first with psychiatric medical management at one site, and substance abuse treatments at another; the second with psychiatric and substance abuse services integrated at one location. All patients will have serial ratings using standard instruments at 0,4,8,and 12 months after hospitalization by an evaluation team blinded to the main study hypotheses and independent of the clinical treatment. If integrated services are more effective than 'standard' services, then we can expect increased compliance with medication, decreased abuse of drugs and alcohol, and improved psychosocial functioning over the study period as demonstrated by clinical and evaluation team ratings.